This morning, we had a power outage at our home. One of the first things I usually try to determine is whether the outage is local or widespread. If it’s local (perhaps a nearby tree that fell?), I need to make sure that the power company knows about the outage and expect that it might take some time to repair, especially if there’s a storm underway. If it’s a widespread outage, it’s probably better if I leave the power company alone because they already know about it.
One of the more interesting uses for Twitter is for sharing information about emergencies. A number of municipalities and news media have Twitter feeds for this purpose. But this uses Twitter as a broadcast medium, rather than enabling users to get information directly from other users.
During the outage, I used my cell phone to check Twitter Search for other word of the outage. Here are some of the tweets:
The power is out (down?)
Our power is out. No ETA yet on restoring it & no clue why. This is going to make for an interesting day.
This isn’t particularly helpful. I don’t know whether these tweeters are on the next street, across town, or are talking about a completely different power outage.
A good location add-on to Twitter should have the following characteristics:
- Entirely opt-in on a per-message basis
- Adds modestly to the size of a tweet
- Provides a consistent format for location so that multiple location-based tweets can be correlated or displayed in a mashup
This week a friend of mine, Tim Burks, introduced a location-based service called tmeet.me . It allows a user to create a tinyURL-like link that can be included in a tweet to indicate one’s location. As the name suggests, tmeet.me is targeted around the use case of telling your friends where you are so that they can join you (for coffee, for example). I think it has considerably broader applicability, and meets these requirements well (although I’m not sure about the mashup part — yet!)
Tools like this enable new and interesting applications for Twitter. I wonder what’s next.
…on the bedpost over night? That’s the question that my 9 year-old daughter, Celeste, answered in this year’s science project.
Bedpost with gum
The question, of course, originates from a novelty song, Does Your Chewing Gum Lose It’s Flavor (On The Bedpost Over Night), popularized by Lonnie Donegan in the early 1960s. I’m not sure how the idea of the experiment originated, but we were talking about suggesting it to Mythbusters when it occurred to us that we could just do the experiment.
We started by going to Home Depot for a “bedpost”, and found a table leg that we varnished and mounted to a stand. We chose three flavors of gum: Wrigley’s Juicy Fruit®, Wrigley’s Spearmint®, and Bazooka® bubble gum. Test volunteers chewed a piece of each flavor, which was split with half being stuck to the bedpost and half placed in an airtight container.
The next day, the pieces of gum were retrieved from the bedpost and the container and arranged in a randomized fashion on a piece of wax paper. The test subjects then rechewed each half, and commented on the flavor of each.
While there was considerable variance in the results, the pieces perceived to have more flavor were evenly divided between the bedpost and the container groups. The flavor differences were generally perceived to be slight.
It’s a little hard to explain standard deviation and confidence levels at the fourth-grade level, but Celeste did understand that the results she got are not sufficient to prove her hypothesis; she would need a lot more data to do that. I was rather surprised that it wasn’t easier to distinguish the bedpost gum from that stored in a sealed container overnight.
Nevertheless, we don’t recommend leaving your gum on the bedpost overnight. It’s very hard to get off. We have considerable experience with that now, having had to clean the bedpost between experiments!
This past weekend, the United States changed over to Daylight Saving Time. Overall, I like DST, even though it made my 6:15 trip to the YMCA this morning very dark.
When Daylight Saving Time was adopted, initially in the early 20th century and then standardized in the 1950s and 1960s, it was a different world. We didn’t have nearly the amount of real-time communication, or even the amount of international air travel that we have today. It really didn’t matter that much when the United States went to Daylight Saving Time relative to the rest of the world.
In the 1980s, the company I worked for subscribed to a bimonthly publication known as the Official Airline Guide (OAG). The OAG listed nearly every scheduled commercial flight in the world and, in the days before the Web, was an essential tool for people like myself who wanted to plan their own air travel. The international edition of the OAG grew to nearly double its normal size every spring and fall. The reason? The US and Europe changed their clocks on different days, about one week apart.
Since that time, we have adjusted the dates for Daylight Saving Time several times, but have never picked the same dates as Europe. It seems like the US likes to remind everyone of its power and sovereignty by making decisions that are different from the rest of the world. After all, as a big and powerful country we have that right, don’t we?
What we ignore in this increasingly global economy is the cost we incur from these decisions. While it’s true that we aren’t killing as many trees on thicker OAGs through the use of the Web, we are instead wasting enormous amounts of administrative time rescheduling meetings for a short period each spring and fall, in addition to lost productivity from missed meetings and the like. DST doesn’t make sense for all countries, and DST has to happen at the opposite time of year in the Southern Hemisphere, so some rescheduling of meetings is inevitable if we are to have DST at all. Of course, things in progress during the time change such as overnight flights will still need special-case schedules on the day of the change.
But why not standardize DST for the Northern Hemisphere? We can start by adopting the European DST schedule in the US. With an increasingly global mindset in Washington, it seems like the time (ahem) is right.
For reference, here’s a handy table of Daylight Saving Time changeover dates.
Image from decade_null via Flickr
Last summer, an Irish cabbie illustrated for me one of the unintended consequences of the use of RFID.
My family and I had just arrived in Dublin, and we were taking a taxi from the airport to the Citywest Hotel, where the 72nd IETF meeting was being held. This involved driving on the M50, a toll road. In the midst of heavy traffic, as we approached a toll barrier, the cabbie (while driving) reached over, opened the glove compartment, and removed a shielded antistatic bag containing a transponder. He attached the transponder to his windshield went through the toll barrier with the transponder allowing him to be billed for the charge.
I asked the driver why he didn’t just keep the transponder on the window. He explained that the transponders are used for more than just paying for tolls in Dublin. They’re also used to pay for parking charges at garages. Someone he knew had the misfortune of driving past a garage exit just as a car was coming out, and the passer-by’s transponder was read by mistake and charged for the exiting car’s parking. So this cabbie keeps his transponder “turned off” by putting it in a shielded bag in his glove box whenever he’s not actually using it. Pretty inconvenient.
Why not just put an on/off switch on each transponder? There is the question of cost, of course, but the bigger issue is that the transponders are bought by the toll collection authorities. If it was easy to turn the thing off, undoubtedly some fraction of people would forget, and the enforcement burden would rise (or the collection rate would fall) as a result.
The real problem here is mission creep. The transponders are being used in applications for which they weren’t designed. There aren’t many passers-by near a toll booth like there are near a garage exit. Perhaps there should be a button on the transponder to permit it to be used for any of these secondary applications. But of course the transponder has no way to distinguish its primary mission (road tolls) from secondary applications such as garage fees.
RFID has lots of interesting applications, and lots of problems too. The security and privacy concerns with RFID are well documented, but this is an issue I hadn’t considered. It was interesting to see how much a cabbie in Dublin knows about it.
A year or so ago, having tried compact fluorescent lighting for a while, I decided it was time to give LED lighting a try.
Most of the lighting in our home is in the form of R30 flood lights in recessed fixtures in the ceiling. We have replaced some of the more frequently-used lights with R30 fluorescent fixtures, and many of you are probably familiar with the slow turn-on time and similar issues.
I ordered three TCPLR30WH27K bulbs in January of last year. The order was delayed several times; apparently the manufacturer (TCP) was having some production problems. They finally arrived in the early summer; we have them installed in the kitchen over the sink and adjacent counter. Each “bulb” is actually an array of lots of individual LEDs; the outer shell is plastic and the whole thing is much lighter than other flood lights.
Here’s a run-down of the pros and cons:
Advantages:
- Instant startup: no wait of several minutes like fluorescents
- Extremely efficient: 6 watts is comparable to a 15W fluorescent or 65W incandescent
- Long life: estimated at 50,000 hours MTBF, vs. 6000 hours for a fluorescent and 1000 hours for an incandescent. 50,000 hours works out to about 27 years of use at 5 hours daily.
Disadvantages:
- Bluish light: This light is rated at 2700 degrees Kelvin, which would be equivalent to a conventional (non-soft white) incandescent light. It’s not as blue as many LEDs I have seen, but don’t let anyone try to tell you this is the same color.
- Flashing: If you’re sensitive to 60 Hz flashing from fluorescent lamps, this is worse. Fluorescents at least have the persistence of the phosphor to mitigate the 60 Hz line frequency, but these are worse. I sometimes see strobe effects in the water coming out of the faucet at the kitchen sink.
- Light pattern: Although the LR30WH27K is classified as a “flood”, I would almost call it a spot. As with many LEDs, the individual LEDs in the bulb focus their light in one direction. Unfortunately, all of the LEDs making up this unit are coplanar; if they had been mounted on a convex surface, the lighting would be much more uniform.
- Cost: LED floodlights are still quite expensive, 3-4 times as expensive as fluorescents. Still, their long life may make this a good deal in hard-to-reach environments.
- Dimmability: This bulb apparently isn’t dimmable, although I understand that there are LED floodlights that are.
Another possible advantage is that we thought we were seeing some bleaching of the color of our kitchen cabinets while the fluorescent floods were installed. I don’t have any data to support this, but we speculated that the fluorescent floods had more UV output than the incandescent floods they replaced. I’m not sure how the LED floods compare, but the fact that their output is more focused causes them to illuminate the cabinets less.
Despite a short list of advantages and longer list of disadvantages, we’re generally happy with the TCP LED floods in this particular application. Most of the lights in our home have dimmers and these bulbs wouldn’t be suitable unless we made them non-dimmable, but in the kitchen the slightly harsh and unconventional light from the LED floods works well.
I have decided that it’s time to start an “external” blog. For some time now, I have maintained blogs on the internal network at work, and will continue to blog there on work-related topics. But from time to time I have wanted to say something to a wider community. Every so often I have wanted to refer a friend to something I have written, and I have one of those “Oh, wait, you don’t have access to that” moments.
There’s more about my intent for this blog on the About page, which I invite you to read and I’ll be updating as my intent becomes more clear and evolves.
I hope this is enjoyable and/or useful.
Altmode 